| Every FLOP Counts: Scaling a 300B Mixture-of-Experts LING LLM without Premium GPUs | ✓ Link | 69.78 | Ling-3.1-flash | 2025-03-07 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 67.6 | Qwen3.5-397B-A17B | 2026-02-16 |
| Rufus-Air: An Open LLM Post-Training Recipe | | 65.8 | Rufus-Air | 2026-09-24 |
| Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning | ✓ Link | 63.8 | Nemotron 3 Ultra | 2026-06-03 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 61.5 | Qwen3.5-122B-A10B | 2026-02-16 |
| Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters | ✓ Link | 60.8 | Step-3.5-Flash PaCoRe | 2026-02-11 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 60.8 | Qwen3.5-27B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 60.0 | Qwen3.5-35B-A3B | 2026-02-16 |
| Introducing OpenAI o3 and o4-mini | ✓ Link | 56.51 | OpenAI o3 | 2025-04-16 |
| Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters | ✓ Link | 55.7 | Step-3.5-Flash | 2026-02-11 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 54.5 | Qwen3.5-9B | 2026-02-16 |
| Kimi K2: Open Agentic Intelligence | ✓ Link | 54.1 | Kimi K2 | 2025-07-28 |
| MAI-Thinking-1: Building a Hill-Climbing Machine | | 53 | MAI-Thinking-1 | 2026-06-02 |
| GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models | ✓ Link | 52.8 | GLM-4.5 (355B) | 2025-08-08 |
| Nanbeige4.1-3B: A Small General Model that Reasons, Aligns, and Acts | ✓ Link | 52.21 | Nanbeige4.1-3B | 2026-02-13 |
| PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning | ✓ Link | 48.0 | PaCoRe-8B (high) | 2026-01-09 |
| Introducing OpenAI o1 | | 44.93 | OpenAI o1 | 2024-09-12 |
| Nanbeige4-3B Technical Report: Exploring the Frontier of Small Language Models | ✓ Link | 41.2 | Nanbeige4-3B-Thinking-2511 | 2025-12-06 |
| From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix | | 37.4 | T-pro-2.1 public | 2026-09-01 |
| LSRIF: Logic-Structured Reinforcement Learning for Instruction Following | | 32.3 | Qwen3-8B (LsrIF) | 2026-01-10 |
| MiniCPM5-1B | ✓ Link | 19.48 | MiniCPM5-1B (Thinking) | 2026-05-25 |